gc.enable()
parameters = parser.parse_args()
min_frac, max_frac, interval, nSamples, edges_csv, delim_option, output, dataset, seq_samples_scale = parameters.min, \
parameters.max, parameters.interval, parameters.samples, parameters.csv, parameters.delim, parameters.output, parameters.dataset, parameters.seq_samples_scale#,\
#[float(x) for x in parameters.eps.split(',')], parameters.title
delim_dict = {0 : '\t', 1 : ' ', 2 : ','}
delimeter = delim_dict[delim_option]
results_dir = 'experiments/results/'
bUndirected = True if parameters.undirected == 1 else False
bfs_method = 'seq'
start_time = time()
print "creating link-server object"
if delimeter == "\t":
print "delimeter is tab"
L=LinkServerCP(dataset, edges_csv, create_new=True, prob_method=parameters.prob_method, prob=[0.1,0.01], delim=delimeter, undirected = bUndirected)
n = L.getNumNodes()
#n = len(V)
print "Number of nodes: ", n
nBFS_samples = 1000
running_times_file = results_dir + output + '-running_times-ratios_k_min-%.3f-k_max-%.3f-samples-%d-bfs_samples-%d-large'%(min_frac,max_frac,nSamples,nBFS_samples)
running_times_file_raw = running_times_file + "-raw"
removeFile(running_times_file)
f = open(running_times_file, 'w')
f_raw = open(running_times_file_raw,'w')
nBFS_samples = 1000
nBFS_samples_theoretic = n * log(n,2)
for nSeeds in xrange(int(min_frac * n), int(max_frac * n), int(interval * n)):
print "k = ", nSeeds
seeds_fname = output + "-seeds-" + str(nSeeds) + '.cp'
runtimes_approx, runtimes_seq = [], []
parameters = parser.parse_args()
min_frac, max_frac, interval, nSamples, edges_csv, delim_option, output, dataset = parameters.min, \
parameters.max, parameters.interval, parameters.samples, parameters.csv, parameters.delim, parameters.output, parameters.dataset
print "Input file: ", edges_csv
print "Output file prefix: ", output
delimiter = delim_dict[delim_option]
start_time = time()
bfs_method = 'seq'
print "creating link-server object"
if parameters.prob_method == 0:
edge_prob = 0.2
else:
edge_prob = [0.1,0.01]
L=LinkServerCP(dataset, edges_csv, create_new=True, prob_method=parameters.prob_method, prob=edge_prob, delim=delimiter, undirected=parameters.undirected)
print "n = ", L.getNumNodes()
V = LoadNodesFromFile(edges_csv, delimiter)
n = len(V)
print 'min_frac', min_frac
k_min = int(n * min_frac)
k_max = int(n * max_frac) + 1
k_step = int(n * interval)
generateSeedFiles(k_min, k_max, k_step, V, nSamples, 'experiments/' + output + "-seeds-")
mean_errors, std_errors = [], []
for k in xrange(k_min, k_max,k_step):
parameters = parser.parse_args()
min_samples, max_samples, samples_step, k_mode, k, edges_csv, delim_option, output, dataset, prob_method, cores= \
parameters.min_samples, parameters.max_samples, parameters.samples_step, parameters.k_mode, parameters.k,\
parameters.csv, parameters.delim, parameters.output, parameters.dataset, parameters.prob_method, parameters.cores
print "Dataset: ", dataset
delim_dict = {0 : '\t', 1 : ' ', 2 : ','}
delimeter = delim_dict[delim_option]
results_dir = 'experiments/results/influence_values/'
bUndirected = True if parameters.undirected == 1 else False
bfs_method = 'seq'
start_time = time()
print "creating link-server object"
if delimeter == "\t":
print "delimeter is tab"
L=LinkServerCP(dataset, edges_csv, create_new=True, prob_method=parameters.prob_method, prob=[0.1,0.01], delim=delimeter, undirected = bUndirected)
n = L.getNumNodes()
if k_mode == 0:
nSeeds = int(n * k)
else:
nSeeds = int(k)
perf_csv_fname = "perf_out" + str(sample(range(1000),1)[0]) + ".csv"
print "Number of nodes: ", n
print "nSeeds = ", nSeeds
results_file = results_dir + output + '-influence_values_samples_min-%d-samples_max-%d-k-%.3f-prob_method-%d'%(min_samples,max_samples, k, prob_method)
removeFile(results_file)
seeds_fname = "%s-seeds-%d.cp"%(dataset,nSeeds)
#!/usr/bin/python
from link_server import LinkServerCP
import sys
import argparse
parser = argparse.ArgumentParser()
parser.add_argument('-cp', type=str)
if __name__ == "__main__":
parameters = parser.parse_args()
print "name of dataset file: ", parameters.cp
L = LinkServerCP(parameters.cp)
new = True
total_time_start = time()
iters = 0
if pr_method == 2:
prob = [0.01, 0.1]
print "min_frac = %.5f, max_frac = %.5f, interval = %.5f" % (
min_frac, max_frac, interval)
print "Number of cores to use: ", parameters.cores
print "CSV file to be used: ", edges_csv
print "creating link-server object"
if delimeter == "\t":
print "delimeter is tab"
L = LinkServerCP(dataset,
edges_csv,
new,
pr_method,
prob,
delim=delimeter,
undirected=parameters.undirected)
V = LoadNodesFromFile(edges_csv, delimeter)
n = len(V)
print "Number of nodes: ", n
running_times_file = output + '-running_times-large-samples-%d-k_frac-%.5f-%.5f' % (
nSamples, min_frac, max_frac)
removeFile(running_times_file)
dRuntimes = defaultdict(list)
for nSeeds in xrange(int(min_frac * n), int(max_frac * n),
int(interval * n)):
start_time_k = time()
n, parameters.graph_method)
running_times_files = [
output +
'-running_times-n-%d-graph_method-%d-samples-prob_method-%d-%d-eps-%.5f'
%
(n, parameters.graph_method, nSamples, parameters.prob_method, eps)
for eps in eps_list
]
for fname in running_times_files:
removeFile(fname)
createRandomGraph(csv_fname, parameters.graph_method, n,
parameters.er_p, parameters.ws_beta, parameters.ws_k)
L = LinkServerCP('input/datasets/' + dataset,
csv_fname,
create_new=True,
prob_method=parameters.prob_method,
prob=[0.1, 0.01],
delim='\t',
undirected=1)
# record loading time of link-server -- for interpolation
removeFile(perf_output_fname)
subprocess.Popen("perf stat -x, -o %s python load_link_server.py -cp %s"%\
(perf_output_fname, "input/datasets/" + dataset), shell = True, stdout = subprocess.PIPE).stdout.read()
nCycles_link_server = getNumCycles(perf_output_fname)
removeFile(perf_output_fname)
#removeFile(csv_fname)
V = xrange(n)
for nSeeds in xrange(int(min_frac * n), int(max_frac * n),
def plot_k_vs_approx():
csv_files = [#'input/datasets/wiki-Vote-small.txt',\
#'input/datasets/email-Enron.txt',\
'input/datasets/Epinions1.txt',\
'input/datasets/Wiki-Vote_stripped.txt']
work_dir = 'experiments/results/'
delims = ['\t', '\t']
n_values = [
LinkServerCP('tmp', csv_fname, create_new=True,
delim=delims[i]).getNumNodes()
for i, csv_fname in enumerate(csv_files)
]
datasets = ['epinions', 'wiki-vote']
x_vals_lists, y_vals_lists, sem_lists = [], [], []
for i, dataset in enumerate(datasets):
print "dataset: ", dataset
n = n_values[i]
min_frac, max_frac, step_frac = 0.01, 0.6, 0.02
k_min, k_max, step = int(n * min_frac), int(n * max_frac) + 1, int(
n * step_frac)
approx_files = [
file for file in os.listdir(work_dir)
if file.startswith(dataset + "-approx_errors-k_min-%d-k_max-%d" %
(k_min, k_max))
]
seq_files = [
file for file in os.listdir(work_dir)
if file.startswith(dataset + "-seq-approx-errors-k")
]
x_vals, y_data, errs = [], [], []
for k in range(k_min, k_max, step):
print "k = ", k
offsets = range(-30, 30)
candidate_fnames = [
dataset + "-approx_errors-k_min-%d-k_max-%d-k-%d-samples-10" %
(k_min, k_max, k + offset) for offset in offsets
]
files_exist = [
candidate in approx_files for candidate in candidate_fnames
]
assert any(
files_exist), "candidate fnames: " + str(candidate_fnames)
offset_idx = files_exist.index(True)
approx_fname = candidate_fnames[offset_idx]
seq_fname = dataset + "-seq-approx-errors-k-" + str(
k + offsets[offset_idx]) + "-samples-10"
print "approx_fname: ", approx_fname
print "seq_fname: ", seq_fname
k, mean_err, err = calculateMeanErrorAndSEM(
work_dir + approx_fname, work_dir + seq_fname)
x_vals.append(1.0 * k / n)
y_data.append(mean_err)
errs.append(err)
x_vals_lists.append(x_vals)
y_vals_lists.append(y_data)
sem_lists.append(errs)
assert len(x_vals) == len(y_data)
assert len(x_vals) == len(errs)
plot2d(x_vals_lists, y_vals_lists, sem_lists, datasets,
[r'$k/n$', 'Approximation ratio'], '',
'experiments/figures/k_vs_approx_combined.pdf')
print "Printing separate plots: "
for i, dataset in enumerate(datasets):
x_data = x_vals_lists[i]
y_data = y_vals_lists[i]
sem_list = sem_lists[i]
plot2d(x_data, [y_data], [sem_list], [dataset], [r'$k/n$','Approximation ratio'], \
'', 'experiments/figures/' + dataset + '-k_vs_approx.pdf')
print "Input file: ", edges_csv
print "Output file prefix: ", output
delimiter = delim_dict[delim_option]
start_time = time()
bfs_method = 'seq'
print "creating link-server object"
if parameters.prob_method == 0:
edge_prob = 0.2
else:
edge_prob = [0.1, 0.01]
L = LinkServerCP(dataset,
edges_csv,
create_new=True,
prob_method=parameters.prob_method,
prob=edge_prob,
delim=delimiter,
undirected=parameters.undirected)
print "n = ", L.getNumNodes()
V = LoadNodesFromFile(edges_csv, delimiter)
n = len(V)
k_min = int(n * min_frac)
k_max = int(n * max_frac) + 1
k_step = int(n * interval)
print "max_k = ", max_frac
print "Minimum k value: %d, maximum k value: %d" % (k_min, k_max)
removeFile(dataset)
generateSeedFiles(k_min, k_max, k_step, V, nSamples,
'experiments/' + output + "-seeds-")
parameters.max, parameters.interval, parameters.samples, parameters.csv, parameters.delim, parameters.output, parameters.dataset#,\
#[float(x) for x in parameters.eps.split(',')], parameters.title
seed()
perf_fname = "runtimes" + str(random())
delim_dict = {0: '\t', 1: ' ', 2: ','}
delimeter = delim_dict[delim_option]
results_dir = 'experiments/results/'
start_time = time()
bfs_method = 'seq'
print "creating link-server object"
if delimeter == "\t":
print "delimeter is tab"
L = LinkServerCP(dataset,
edges_csv,
create_new=True,
prob_method=2,
prob=[0.1, 0.01],
delim=delimeter)
V = LoadNodesFromFile(edges_csv, delimeter)
n = len(V)
print "Number of nodes: ", n
running_times_file = results_dir + output + '-running_times-ratios_k_min-%.3f-k_max-%.3f-samples-%d' % (
min_frac, max_frac, nSamples)
running_times_file_raw = running_times_file + "-raw"
removeFile(running_times_file)
f = open(running_times_file, 'w')
f_raw = open(running_times_file_raw, 'w')
for nSeeds in xrange(int(min_frac * n), int(max_frac * n),
int(interval * n)):
print "k = ", nSeeds
seeds_fname = output + "-seeds-" + str(nSeeds) + '.cp'
def print_out(text):
if output_mode in [0, 2]:
print text
if output_mode in [1, 2]:
f_output = open(output_file, 'a')
f_output.write(text + '\n')
f_output.close()
if __name__ == "__main__":
parameters = parser.parse_args()
dataset, seeds_file, results_file, output_mode, nSamples = parameters.dataset, parameters.seeds_file, parameters.results_file,\
parameters.output_mode, parameters.nSamples
L = LinkServerCP(dataset)
f = open(seeds_file, 'r')
assert (nSamples <= 0 or parameters.min_relative_standard_error <= 0)
seeds_list = cp.load(f)
f.close()
removeFile(parameters.reached_nodes_file)
if output_mode in [1, 2]:
f = open(results_file, 'w')
for i, seed_set in enumerate(seeds_list):
if output_mode in [0, 2]:
print "Sample ", i
if parameters.get_n_reached == 1:
try:
avg, total_samples,l_n_reached = sequential_estimation(L, seed_set, max_samples_cap=nSamples, \
nCores=parameters.cores,bReturnValues = True,\
